Carry propagation (was: Re: Risc V greatly underperforms)

2021-10-04 Thread Niels Möller
Torbjörn Granlund writes: > This all means that on Risc V, multi-word subtraction could be made to > at 2 cycles/word while multi-word addition is limited to 3 cycles/word, > in both cases assuming a very wide super-scalar core. Remember that > other concurrent CPUs do these in 1+epsilon cycles/

Re: Please update addaddmul_1msb0.asm to support ABI in mingw64

2021-10-04 Thread Niels Möller
Marco Bodrato writes: > While looking at this e-mail on gmp-bugs, I added DOS support and also > reordered the branches around the exit code. > Do you agree with those changes? Looks reasonable to me. To be honest, it's a bit difficult to follow, it seems I didn't document this that well when I

Re: Please update addaddmul_1msb0.asm to support ABI in mingw64

2021-10-04 Thread Marco Bodrato
Ciao Niels, you are the author of this code. Il 2020-06-07 21:48 Idigger Lee ha scritto: Please update addaddmul_1msb0.asm to support ABI in mingw64. While looking at this e-mail on gmp-bugs, I added DOS support and also reordered the branches around the exit code. Do you agree with those c